1,039 open questions on Ugandan culture, history and society across 24 domains (education, traditional medicine, media, economy, literature, social norms and more), from the Uganda Cultural and Cognitive Benchmark, imported from inspect_evals. The model answers in free text under the eval's own system message. An LLM judge compares each reply with the reference answer using the eval's own grading prompt, with every judge call routed and metered through our gateway. Cases are served in a fixed seeded shuffle (the eval shuffles per run). Benchmark score = plain accuracy over the cases run. The eval grades with GPT-4o-mini; this Benchmark pins our house judge. inspect_evals runs the model at temperature 0 with 512 max tokens; this benchmark imposes no generation settings, so long replies are judged in full. Non-commercial licence. Answer format: free text, no fixed format. A judge compares the whole reply with the reference answer: it scores 1 when it carries the reference's main facts, in any wording and with extra relevant detail allowed, and 0 when it gets Ugandan culture or history wrong, misses the key points, or describes another culture.